https://github.com/snap-stanford/gib

Graph Information Bottleneck (GIB) for learning minimal sufficient structural and feature information using GNNs
https://github.com/snap-stanford/gib

graph-neural-networks information-bottleneck node-classification representation-learning robustness

Last synced: 3 months ago
JSON representation

Graph Information Bottleneck (GIB) for learning minimal sufficient structural and feature information using GNNs

• Host: GitHub
• URL: https://github.com/snap-stanford/gib
• Owner: snap-stanford
• License: mit
• Created: 2020-06-11T02:29:19.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2022-11-29T05:51:44.000Z (almost 2 years ago)
• Last Synced: 2024-06-29T07:52:26.348Z (4 months ago)
• Topics: graph-neural-networks, information-bottleneck, node-classification, representation-learning, robustness
• Language: Jupyter Notebook
• Homepage:
• Size: 211 KB
• Stars: 120
• Watchers: 46
• Forks: 24
• Open Issues: 1
• Metadata Files:
  • Readme: README.rst
  • License: LICENSE

Awesome Lists containing this project

• StarryDivineSky - snap-stanford/gib - Cat 和 GIB-Bern，二者在抵御对抗攻击时取得了优异的性能。 图信息Bottleneck打造图最优表示->避免过拟合，并具备稳健性 (图对抗攻击 / 网络服务_其他)